In [5]:
# Create new dataset that sums: total vaccinations, daily vaccination, and population by date and country group.
# Call this new dataset that is by country group simply by owid
owid = owid_merged.groupby(['country_group', 'date']).agg({'total_vaccinations':'sum', 'daily_vaccinations_raw':'sum', 'daily_vaccinations':'sum', 'population':'sum'}).reset_index()
# Create variable that is the cumulative sum of daily vaccinations.
owid['cum_daily_vacc'] = owid.groupby(['country_group'])['daily_vaccinations'].cumsum()
# Create new column with total_vaccinations per hundred
owid['total_vaccinations_per_hundred'] = (owid['cum_daily_vacc']/owid['population'])*(100)
# Drop obs before 15th december 2020. This is because only Advanced Economies had started.
owid = owid[~(owid['date'] < '2020-12-15')]
# Prepare data for plotting.
owid = owid.set_index('date')
# Create separate dataframes.
owid_advanced = owid[owid.country_group == "Advanced Economies"]
owid_emde = owid[owid.country_group == "EMDEs"]
# Do a 30-day moving average of total vaccivations per hundred
#owid_advanced['ma_total_vaccinations_per_hundred'] = owid_advanced['total_vaccinations_per_hundred'].rolling(7).mean()
#owid_emde['ma_total_vaccinations_per_hundred'] = owid_emde['total_vaccinations_per_hundred'].rolling(7).mean()
x = np.array(owid_advanced.index.values)
y1 = np.array(owid_advanced['total_vaccinations_per_hundred'])
y2 = np.array(owid_emde['total_vaccinations_per_hundred'])
plt.plot(x, y1, "-b", label="Advanced Economies")
plt.plot(x, y2, "-r", label="EMDEs")
plt.legend(loc="upper left")
plt.ylim(0, 140)
plt.title("Total vaccinations per hundred")
plt.savefig("total_vaccinations_per_hundred.png")
plt.show()
